Toward a rhetoric of informating texts

This paper begins by asking why hypertext researchers publish their work in print and compose their hypertexts from previously printed sources. It argues that these practices limit the development of hypertext rhetoric by privileging a discrete or granular model of discourse nodes as stable units connected by purely transitional links. The paper explores the limits of the node/link model, suggesting that links can take on certain properties of nodes and vice versa. Drawing on the phenomenological critique of rationalist mechanism developed by Winograd and Flores, the paper presents an alternative discourse model for hypertext which regards nodes and links in complementarily, as contingent structures subject to conceptual “breakdown.” Applying this model to actual communication practices, the paper invokes Zuboff’s distinction between “automating” and “informating” applications of technology, outlining a rhetoric based on a constantly evolving textual structure in which object relations remain fluid. A new term is proposed, the informand, to designate the communal, interactive discursive space created by informating systems like hypertext and artificial realities. The paper concludes by urging experimentation with informating practices in hypertext, a move away from print models and toward all-electronic composition. Permission to copy without fee alt or part of this material is granted provided that copies are not made or distributed for direct commercial advantage, the ACM copyright notice and the title of the publication and its date appear, and notice is given that copying is by permission of the Association for Computing Machinery. To copy otherwise, or to republish, requks a fee and/or specific permission. @1992 ACM O-89791-547-X/92/0011 /0171/ $1.50 1 Where are the hypertext? At the first European Conference on Hypertext, the theorist and developer Mark Bernstein asked, where are the hyperfexts ? Then as now, one could point to a number of experimental ventures; but important as they are, these examples do not sufficiently answer Bernstein’s challenge. If hypertext equals print in importance and utility, then its advocates should be able to adduce many practical applications in regular use, especially in technical fields. Yet relatively few substantial, long-term applications of hypertext have appeared so far. Hypertext has yet to become the primary medium in any commercial or intellectual community, Even in hypertext research and development itself, progress has been limited. Leaders in the field have produced groundbreaking hypertexts-onhypertext, but major conferences and research organs do not yet accept hypertextual submissions. What you are now reading, after all, is a traditionally structured, linearly argumentative theoretical paper. Why is this so? To ask a variant of Bernstein’s question: why isn’t this paper a hypertext? This is a little like questioning the emperor’s fashion sense — dangerous, because such questions tend to breed. Corollaries here might include the following: Why does the hypertext research community publish its work in print? Do we stay with this medium simply out of institutional inertia, or does print provide functionality that current hypertext systems do not? Why would we want to write hyperdocuments instead of linear papers? How would this shift in media affect audience, address, and message structure; and how would these changes relate to our practices as a research community? 172 ACM ECHT CONFERENCE These questions belong to the domain of hypertextual rhetoric — a field which is not very well prepared to address them. Much of the research in this area so far has assumed a close congruence between hypertext and print, generating rhetorics of “reform” which limit the new medium’s departures from earlier conventions [3; 23]. Present rhetorics assume an individual or one-to-many context of address rather than a collaborative social situation. They also also tend to regard texts as closed and definitive objects delivered in some static medium such as print. Neither of these emphases is very useful for hypertext. This paper proposes a new framework, a rhetoric oj informating texts, which acknowledges that both the discursive properties and the social implications of hypertext may differ substantially from those of print. Bolter [5] has demonstrated the importance of hypertext in the evolution of writing systems. Landow [17] has explored the convergence of hypertext with contemporary theories of interpretation. Both foresee large changes in the contexts for writing as hypermedia systems facilitate changes in authorship, publishing, and intellectual property. To evolve a rhetoric adequate to these changes, we need to extend Landow’s and Bolter’s analyses. Landow finds consonance between hypertext and poststructuralism, which approaches texts as dynamic, polyvocal networks of expression. A similar emphasis can be found in the cognitive theory of Winograd and Flores [32], whose critique of rationalism in the design of information systems describes a tension between static and dynamic structures, formalizations and “breakdowns.” This perspective has considerable value for an approach that differentiates electronic communication from older technologies. In exploring the social implications of this difference, Bolter relates the development of hypertext to a shift away from absolute hierarchies, especially in information work. In her ethnographic studies of management, Zuboff has explored this transformation extensively. Her discussion of “automating” and “informating” strategies in industry [35] provides an important link between the issues of hypertextual design raised by Bolter and their likely impact on the 21st-century workplace, particularly in the area of “industrial-strength hypermedia. ” These concepts suggest possibilities for new theoretical understandings of hypertext; but since rhetoric is a practical field, its contributions cannot be limited to theory alone. The focus of this paper therefore remains Bernstein’s eminently pragmatic question: where are our hypertext? 2 Here are the hypertext Like most academics and professionals, hypertext researchers depend on established channels of dissemination. We produce unified, monological discourses: research reports, theoretical papers, and books. We thus create an apparent inconsistency. How can we insist on the usefulness of hypertext while we communicate mainly in print? Until recently it was possible to plead software and hardware constraints, but this disclaimer seems less plausible now that robust hypertext environments are readily available. A number of researchers regularly use hypertext systems in support of their printed work [4; 6; 21; 26]. More to the point, a large body of technical literature about hypertext is available in hypertextual form. To name three prime examples: the Guide envelope version of Nelson’s Literary Machines [24], the Association for Computing Machinery’s Hyperfexf on Hyperfext [33], and the ACM Hyperfexf Compendium [1]. To a certain extent, these projects answer Bernstein’s query. Researchers have produced at least the beginnings of a hypertextual literature, an experimental base from which conventions about form, procedures, and design will emerge. Yet in a sense the current generation of hypertexts-on-hy pertext provide only a partial response to the demand for practical implementations. They are hypertext, to be sure, but hypertext of only one fairly primitive kind. This type might be called the hypertext retro~it: as in the ACM Compendium, discourse appears first in print and is reprocessed into linked, electronic form. Thus the paper you are MILANO, NOVEMBER 30DECEMBER 4, 1992 173 now reading is at least implicitly hypertextual; indeed, all academic writing would be. Such traditional features as citations and footnotes are proto-hypertextual [181. But print-hypertext conversions represent only one possibility. The Cornpendiu w, for instance, contains links and other navigational devices for intertextual movement, but these mechanisms were added as superstructure. Projects like this are irzcu nabu la or cradle works, reflecting the influence of both old and new media [22]. The history of writing teaches the enormous value of such transitional products; but we must remember that they prefigure further developments.